This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. All malignancies require well-organized metabolic activities to support growth and proliferation. Since tumor metabolism is essential to growth, an understanding of the metabolic properties of a tumor offers the opportunity to 1) identify critical steps that may be exploited in therapeutic development and 2) to develop well-defined noninvasive biomarkers of the status of a tumor and its response to treatment. Although the overall network of metabolic pathways is complex, the fundamental pathways are clearly defined and include oxidation of blood-borne substrates in the citric acid cycle, oxidation of glucose to lactate, and flux through the pentose phosphate pathway. In this Project we will examine metabolic and genetic features of human tumors implanted in the mouse brain as a model of primary and metastatic cancers of the central nervous system. We have already shown that superb-quality 13C NMR spectra can be obtained from tissue extracts of the mouse brain and implanted tumors. Since the brain is composed of both glia (astrocytes and oligodendrocytes) and neurons and the metabolic activities of each cell type may differ, these experiments will use the rich information content of a 13C NMR spectrum plus judicious selection of labeling patterns and substrates to assess fluxes in each compartment separately. Two central questions in cancer metabolism can be addressed. The first is whether the metabolic pathways required for tumor growth in the brain are common among the cancer subtypes or are subtype-specific. For example, the metabolic features of normal breast, skin, lung and kidney are quite distinct, yet it is not known whether these characteristic features are preserved in tumors that derive from these organs and metastasize to the brain. The second question is whether the metabolic phenotype of a malignancy in the brain is a consequence of its interaction with the local microenvironment or, alternatively, is a consequence of the genotype of the primary tumor. Designing therapies as well as biomarkers of response based on metabolism hinges on understanding these fundamental properties of cancer cells. To address these questions, we will use our human orthotopic brain tumor mouse models, which include glioblastoma (GBM) and the four most common tumors which metastasize to the brain (melanoma, lung, breast and renal cell cancer) which were derived from patients at the time of tumor resection, have been passaged exclusively in mouse brain, and are fully characterized molecularly. They provide an unprecedented opportunity to compare cancer subtypes directly in a common microenvironment using 13C NMR spectroscopy to measure relative fluxes in the pentose phosphate pathway, the citric acid cycle and glycolysis. There are three aims: Aim 1. To measure the metabolic phenotype of glioblastoma. Presumably these tumors are derived from mature glial cells with the implication that the two characteristics of glial metabolism (compared to neurons) are preserved: oxidation of acetate and absence of production of GABA from the citric acid cycle. Since gliomas grow in the environment of neurons, metabolism of tracers by neurons must be separated from metabolism of glial cells. The relative rates of glucose and acetate oxidation and generation of GABA from the citric acid cycle will be measured by 13C NMR isotopomer analysis of the malignancy during metabolism of 13C-enriched acetate, 13C enriched glucose, or both. Aim 2. To examine intermediary metabolism in brain metastases and determine the variation in pathway activity based on histological subtype. The four most common types of brain metastases will be used for these studies. These include non-small cell lung cancer, breast, renal cell, melanoma. Aim 3. To compare and contrast the metabolic phenotype of brain metastases with GBM and determine the impact of the underlying status of the common cancer pathways, RAS-MAPK, AKT, and p53 pathways relative to the underlying histology. We will determine whether the genotype or cell type is more important in determining the metabolic phenotype.